TIMAP Upregulation Correlates Negatively with Survival in HER2- Negative Subtypes of Breast Cancer

Objective: TIMAP expression is regulated by transforming growth factor beta 1 (TGFβ1); known for its role in breast cancer development and metastasis. Nevertheless, data on TIMAP protein expression and its association with breast cancer development are lacking. In this study, we aimed to investigate the variation in TIMAP protein expression in breast cancer tissue and its correlation with various clinicopathological characteristics of breast cancer patients and overall survival rate. Methods: A total of 159 paraffin-embedded tissue blocks from women diagnosed with four breast cancer subtypes (49 HER2-only, 33 Luminal A, 39 Luminal B, and 38 triple negative) were used to construct tissue microarray (TMA), followed by TIMAP immunohistochemistry (IHC). TIMAP expression was scored by two pathologists and categorized as weak (1-33% expression), moderate (34-66%), and strong (67-100%). Chi-square test and Kaplan Meier survival test were performed to determine the association between TIMAP expression and clinicopathological features and overall survival rate, respectively. Results: TIMAP protein was strongly expressed in 46 (93.9%) HER2-only, 32 (97%) luminal A, 37 (94.9%) luminal B, and 29 (76.3%) triple negative. TIMAP expression negatively associated with ER/PR expression (P=0.03), and it negatively impacted the overall survival in HER2 negative group (P=0.02). Conclusion: Our findings suggest that TIMAP protein expression is upregulated in all breast cancer subtypes. However, its prognostic role is exclusively observed in HER2- negative group, suggesting a potential of targeting TIMAP in future therapeutic strategies in this group.


Introduction
TIMAP (TGFβ-Inhibited Membrane-Associated Protein); or as annotated in GenBank PPP1R16B, is a regulatory subunit of protein phosphatase 1 catalytic subunit (PP1c) predominantly expressed in central nervous system (CNS) and hematopoietic cell lines (Cao et al., 2002;Magdaleno et al., 2002). It regulates PP1c enzymatic activity by targeting the holoenzyme to specific subcellular locations and substrates; subsequently affecting target signal transduction cascades involved in controlling various cell processes (Boratkó and Csortos, 2017). For example, it binds and regulates the phosphorylation state of the non-integrin laminin receptor1 (LAMR1) (Kim et al., 2005); best known for its association with tumor progression, metastasis (Fülöp and Larbi, 2002), and tumor angiogenesis (Iwamoto et al., 1996). In addition, TIMAP interacts with and regulates the function of many other key molecules involved in controlling cell adhesion, survival, proliferation and migration. Among these molecules are RESEARCH ARTICLE

TIMAP Upregulation Correlates Negatively with Survival in HER2-Negative Subtypes of Breast Cancer
myosin light chain 2 (MLC2) (Shopik et al., 2013;Wang et al., 2019) and Ezrin-Radixin-Moesin (ERM) family (Csortos et al., 2008;Boratkó and Csortos, 2017) that stimulate dynamic remodeling of cell cytoskeleton during cell adhesion, proliferation and migration. Another example is the eukaryotic elongation factor-1A (eEF-1A) (Boratkó et al., 2015), which exhibits roles in cytoskeleton organization, apoptosis, and regulation of protein expression (Sasikumar et al., 2012). Moreover, TIMAP has been shown to regulate the phosphorylation level and subcellular localization of merlin (Boratkó et al., 2017); a tumor suppressor known for regulating cell proliferation (Morrison et al., 2001) and organizing cellular junctions (McClatchey and Fehon, 2009). Also, TIMAP inhibits the tumor suppressor (PTEN)-mediated suppression of Akt activity in human endothelial cells (Obeidat et al., 2014). Akt activation is key to promoting cell survival, proliferation, angiogenesis, and tumor progression. On the other hand, PTEN antagonizes Akt effects, and thus is inhibited in many tumors. Altogether, these studies suggest that TIMAP might play an important role in cancer. TIMAP is down regulated by transforming growth factor beta (TGFβ1) (Cao et al., 2002;Yang et al., 2017) which is well known for its role in regulating various cellular processes, including proliferation, apoptosis, migration, angiogenesis, differentiation, inflammation, and cell cytoskeletal remodeling (Clark and Coker, 1998). Furthermore, TGF-β is a key player in cancer development, and it is the main stimulator of epithelialto-mesenchymal transition (EMT); a cellular process that potentiates cancer cell metastasis (Garcia et al., 2018). TGFβ is an important regulator of normal mammary gland development, as well as breast cancer development and progression (Burdette et al., 2005), where it plays a dual role (Huang et al., 2018); while in early stages of mammary carcinogenesis it acts as a tumor suppressor, in late stages it promotes tumor progression by potentiating tumor metastasis (Muraoka et al., 2002;Nam et al., 2008). Hence, it's plausible to hypothesize that TIMAP expression might be altered in breast cancer.
Large scale mutagenesis analyses in mice sensitized to c-Myc-induced apoptosis identified TIMAP as one of key cellular oncogenes (Mendrysa et al., 2010). Moreover, according to RNA sequencing data, TIMAP expression appears to be altered in different solid tumors, including breast cancer, and hematological malignancies [https:// www.proteinatlas.org/ENSG00000101445-PPP1R16B/ pathology]. These data also indicated that it acts as a prognostic biomarker in head and neck and cervical cancers. However, to date, TIMAP expression at protein level and its role in breast cancer, or any kind of cancer for that matter, have not been studied.
In the current study, we investigated TIMAP protein expression in 159 breast cancer tissue samples categorized into the following subtypes; Human epidermal growth factor receptor 2 (HER2)-only, Luminal A, Luminal B, and triple negative, by immunohistochemistry (IHC) using tissue microarray (TMA) and correlated its protein expression with different clinicopathological features and overall survival rate.

Materials and Methods
Tissue specimens. A total of 159 women diagnosed with various breast cancer subtypes were identified via retrospective analysis of the pathological records from the department of pathology at King Abdullah University Hospital (Irbid, Jordan). The patients had been treated with radical mastectomy for the tumor or axillary lymph node resection between the years 2007 and 2019. This study was approved by the Faculty of Medicine Research Ethics Committee at Jordan University of Science and Technology (Irbid, Jordan).
TMA and IHC. Archived paraffin-embedded breast carcinoma tissue blocks were used to construct the TMA using TMA Master II instrument (3DHISTECH Ltd., Budapest, Hungary). TMA tissue blocks were sectioned at 4 μm thickness and collected on Superfrost plus glass slides for processing by IHC using the BenchMark ULTRA system (Roche Diagnostics, Risch-Rotkreuz, Switzerland) as previously described (Bodoor et al., 2018;Bodoor et al., 2020). Rabbit polyclonal antibody against TIMAP/PPP1R16B (MyBioSource, Inc, San Diego, USA) at 1:200 dilution was used to determine TIMAP protein expression. The slides were scored as described previously (Bodoor et al., 2018;Bodoor et al., 2020), whereas slides with 1-33% expression were scored as weak and were considered negative in statistical analysis; slides with 34-66% expression were scored as moderate and were considered positive; and slides with 67-100% expression were scored as strong and were also considered as positive in statistical analysis. The expression of Estrogen receptor (ER), Progesterone receptor (PR) and HER2 was retrieved from the patients archived records. Tumor volume was calculated differently according to available data for dimensions; For one-dimension x: volume = 4/3*π*(x/2)3; for two dimensions x and y whereas x<y: volume = x2*y; and for three dimensions x, y and z: volume = x*y*z.
Statistical Analysis. The relationship between TIMAP expression level and clinicopathological characteristics was established by Chi-square test. One-sided Fisher's exact test was considered more reliable for 2×2 crosstabs and thus was reported instead of Chi-square test where appropriate. Survival outcomes, defined as the period from time of diagnosis to death from any cause or the last contact, were estimated with the Kaplan Meier analysis and compared between groups by log-rank test. Statistical analyses were conducted using IBM SPSS Statistics for IOS, Version 26 (IBM Corp, Armonk, NY, USA). p-value ≤0.05 was considered significant.

Results
Patients clinicopathological characteristics and TIMAP expression levels. The clinicopathological characteristics of the 159 study subjects are shown in Table  1. Specimens of four molecular subtypes of breast cancer were included in the study; 49 HER2-only (ER-/ PR-/ HER2+), 33 luminal A (ER+/PR+/HER2-), 39 luminal B (ER+/PR+/HER2+), and 38 triple negative (ER-/PR-/ HER2-). TIMAP expression was scored according to percentage of positive cells (as described in the methods section), and it was categorized into "negative" for both null and weak expression; and "positive" for both moderate and strong expression (Figure 1) Figure 2).
Associations between TIMAP expression and clinicopathological characteristics of patients. Analyses of association between TIMAP expression and clinicopathological features of patients are displayed in Table 2. There was no significant association with most of the features. However, significant associations were observed between TIMAP expression and the molecular subtypes of breast cancer (p=0.007), ER/PR expression (P=0.03), and TIMAP subcellular localization (P=0.000).
( Figure 3A, P=0.3), as well as when each subtype was analysed independently ( Figure 3B (HER2-only); P=0.75, 3C (Luminal B); P=0.6, 3D (Triple negative); P=0.3). However, since TIMAP expression was negatively and It is worth mentioning that there was a positive association between TIMAP expression and HER2 expression (61 cases in HER2 negative group and 83 in HER2 positive), however, it did not reach statistical significance (P=0.06), while TIMAP expression negatively associated with ER/PR expression (75 cases in ER/PR negative group and 69 in ER/PR positive, P=0.03).
Associations between TIMAP expression and survival outcomes. We next investigated the impact of TIMAP expression on overall survival. There was no significant difference between negative and positive TIMAP expression when all subtypes were pooled together  positively correlated with ER/PR and HER2 expression, respectively (Table 2), we further investigated its impact on overall survival in those groups. While there was no significant difference between negative and positive TIMAP expression among ER/PR negative ( Figure

Discussion
Previous transcriptome studies of different cancer cohorts have shown that TIMAP is upregulated in various tumor types, whereas it acts as a prognostic marker in both cervical and head and neck cancers [https:// www.proteinatlas.org/ENSG00000101445-PPP1R16B/ pathology]. However, to date, immunohistochemical data of TIMAP expression in cancer tissue and its clinical significance are lacking. Here, we present a novel evidence of TIMAP protein expression in breast cancer tissue.
In our study, TIMAP expression was significantly augmented in all subtypes of breast cancer compared to normal breast tissue, which showed minimal to no TIMAP expression (Figure 1). This result is consistent with previous studies that demonstrated a hematopoietic and central nervous system (CNS)-predominant expression of TIMAP in normal tissue (Cao et al., 2002;Magdaleno et al., 2002), and that its expression, at least at transcript level,    was upregulated in cancer [https://www.proteinatlas.org/ ENSG00000101445-PPP1R16B/pathology]. However, our study provides an unprecedented evidence of TIMAP upregulated expression at protein level in breast cancer.
Among the four subtypes studied here, TIMAP expression was observed less frequently in triple negative group, which is usually correlated with poor prognosis, while HER2-only, luminal A, and luminal B groups showed relatively similar and higher levels of expression. This result could not be directly related to the lack of expression of ER, PR, or HER2, as the non-triple negative subtypes express at least one of these proteins and show similar expression of TIMAP. However, since TIMAP has been shown to be downregulated by TGFβ1 (Cao et al., 2002;Yang et al., 2017) and that TGFβ1 is significantly upregulated in triple negative breast cancer as opposed to non-triple negative subtypes (Ding et al., 2016), it is plausible to suggest that the reduced level of TIMAP protein observed in the triple negative subtype here could be explained by presumably upregulated TGFβ1 expression. Nevertheless, future analyses are required to examine this possibility.
Furthermore, TIMAP expression was positively correlated with HER2 expression and negatively correlated with ER/PR expression. Whether there is a regulatory relationship between these molecules is still unknown and requires further investigation. Nonetheless, since, overall, the prognosis of ER/PR-negative tumors is relatively poorer than for ER/PR-positive, targeting TIMAP may present a new therapeutic opportunity in treatment of challenging subtypes of breast cancer.
Most breast cancer samples in this study, regardless of their subtypes, displayed both cytoplasmic and nuclear expression patterns of TIMAP. This is in agreement with previous studies that showed TIMAP can be localized to the plasma membrane, cytoplasm and the nucleus (Cao et al., 2002;Kim et al., 2005;Li et al., 2007;Csortos et al., 2008;Boratkó et al., 2015;Boratkó and Csortos, 2017); and this is attributed to its N-terminal nuclear localization signal and the C-terminal CAAX domain (Cao et al., 2002) that allow TIMAP to shuttle between these cell compartments.
It has been previously shown that overexpression of TIMAP induces endothelial cell survival, proliferation and angiogenesis (Obeidat et al., 2014), as well as TGFβ1-stimulated macrophage migration (Yang et al., 2017). Since these cellular processes are indispensable to tumor development and metastasis, we further analyzed whether TIMAP expression level was correlated with tumor development stage and metastasis state, however there was no significant correlation.
Our survival analysis demonstrated a clear negative impact of TIMAP expression in HER2-negative groups that are usually more resistant to treatment. This finding further highlights the potential role of TIMAP as a therapeutic target and, thus, we recommend future investigations to address the molecular mechanism of this protein in breast cancer development and the possibility of targeting it in the treatment of this disease, especially in drug resistant subtypes.

Author Contribution Statement
Marya Obeidat carried out the study design, examined the IHC staining, performed statistical analyses, drafted and prepared the final manuscript. Khaldon Bodoor participated in study design and helped in drafting and reviewing the manuscript. Mohammad Alqudah coordinated data collection and TMA and IHC staining, and pathological review of sections. Amro Masaadeh and Marwa Barukba scored the IHC staining and captured the microscopic images. Rowida Almomani participated in study design.